Question

我正在将Testng断言转换为AssertJ我正在使用sed命令。当有新行如

的陈述时，我被绞死了

Exa 1：

assertEquals(person.getAddressLine1(),
        "This is long line, so needs to go in second line as per Formatter.........!!!!!!!");

Exa 2：

assertEquals(person.getAddressLine1(),
             "This is very long line, so needs to break down in multiple" +
                     "This is very long line, so needs to break down in multiple" +
                     "This is very long line, so needs to break down in multiple");

我当前的解决方案：这对于单行断言非常适合。

sed -i 's/assertEquals([[:blank:]]*\(.\+\),[[:blank:]]*\(["].\+["]\+\))[[:blank:]]*;/Assertions.assertThat(\1).isEqualTo(\2);/g'

注意：我知道，我可以在编辑器Formatterr中更改行的最大长度，但不知何故我无法更改它因为我需要做 Ctrl 再次+ Shift + F ，这对我来说是不可能的。

更新：两个示例的预期输出如下所示。

Exa 1：

Assertions.assertThat(person.getAddressLine1()).isEqualTo("This is long line, so needs to go in second line as per Formatter.........!!!!!!!");

Exa 2：

Assertions.assertThat(person.getAddressLine1()).isEqualTo("This is very long line, so needs to break down in multiple" +
        "This is very long line, so needs to break down in multiple" +
        "This is very long line, so needs to break down in multiple");

Answer 1

sed可能会这样做，但由于存在换行符，命令会更复杂。这是一个更简单的perl命令，使用slurp模式执行相同的操作：

perl -00 -pe 's/\bassertEquals\(\s*([^,]+),\s*([^)]+)\)\s*;/Assertions.assertThat($1).isEqualTo($2);/' file

RegEx Demo

使用sed即可：

sed -i -E ':a;N;s/\n[ \t]*//g;s/;/;\n/;s/assertEquals\([[:blank:]]*([^,]+),[[:blank:]]*([^)]+)\)[[:blank:]]*;/Assertions.assertThat(\1).isEqualTo(\2);/;ba' file

Answer 2

感谢各位的支持，我终于可以根据我的要求使用sed了。

sed -i ':a;N;$!ba;s/[[:blank:]]*assertEquals(\([^;]\+\),\(\n[[:blank:]]*\([^;]\+\(\n\)*\)\))[[:blank:]]*;/    Assertions.assertThat(\1).isEqualTo(\2);/g' '{}' \;

Answer 3

当assertEquals的第一个参数完全在第一行并且不包含逗号时，这是对你想要的函数的简单近似：

sed -i 's/assertEquals(\s*\([^,]\+\),/Assertions.assertThat(\1).isEqualTo(/g'

Answer 4

如果您可以访问awk，也可以

<强>脚本

awk '$0 ~ /assertEquals\(person.getAddressLine1\(\),/{
printf "Assertions.assertThat(person.getAddressLine1()).isEqualTo(";
getline;
gsub(/^[[:blank:]]*"/,"\"")}1' filename

示例输出

Assertions.assertThat(person.getAddressLine1()).isEqualTo("This is very long line, so needs to break down in multiple" +
        "This is very long line, so needs to break down in multiple" +
        "This is very long line, so needs to break down in multiple");

在Unix中使用sed处理新行

4 个答案: